Meta-learning is one of the hottest topics of research. In recent years, many researchers have taken interested in meta-learning due to its promise to reach generalization. In this blog post, we present a brief introduction to concepts in meta-learning and some applications which will immerse you in this wonderful field. An starting point of view One can notice that humans or animals can learn a task quickly when they have prior knowledge on another similar task.

La segunda acepción de la RAE nos dice que una secuencia "es una sucesión de cosas que guardan entre sí cierta relación". En el mundo real podemos encontrar innumerables ejemplos que se ajustan a esta definición. Por ejemplo, en aplicaciónes de transformación de audio a texto y visceverza, pronóstico del clima, en el control de robots o procesamiento de video, etc. En ese sentido, es importante poder abstraer modelos matemáticos que nos permitan estudiar mejor estos procesos.

Reinforcement learning is a very interesting research topic, in recent years, their applications and their performance have been growing quickly. In this blog post, we present the basics of model-based reinforcement learning and its recent advances.